| title | UniRestorer++: Scalable Universal Image Restoration with Multi-Granularity Degradation Modeling |
|---|---|
| authors | Jingbo Lin, Zhilu Zhang, Fan Li, Renjing Pei, Wenbo Li, Hang Xu, Hongzhi Zhang, Wangmeng Zuo |
| date | 2026-11-04 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2611.06042 |
| upvotes | 2 (Hugging Face Daily Papers, 2026-11-06) |
| links | abs pdf hf code |
| cite | |
| abstract | Universal image restoration aims to handle diverse degradations within a single model, but a reliable scaling strategy remains underexplored. Increasing backbone depth and width can lead to performance saturation, while degradation-aware specialization depends on accurate degradation estimation and is vulnerable to errors on out-of-distribution and real-world inputs. We propose UniRestorer++, a scalable framework based on multi-granularity degradation modeling. It progressively refines degradation groups and introduces group-specific experts, coupling capacity growth with specialization. Degradation-estimation-based routing selects a root-to-leaf path, while granularity-estimation-based routing chooses a level along that path to balance specialization and robustness. Spatially dense degradation representation maps further enable region-aware routing for spatially non-uniform corruptions. We develop CDD-50, a benchmark covering 50 degradation categories from single to penta-degradation. Experiments show approximately log-linear PSNR trends over the evaluated model-size range and stronger scaling gains on out-of-distribution and real-world data than vanilla MoE scaling. UniRestorer++ achieves state-of-the-art performance across benchmarks and demonstrates zero-shot generalization and few-shot transfer on real-world images. Code: https://github.com/mrluin/UniRestorer_plusplus |
bibtex
@misc{lin2026unirestorer,
title={UniRestorer++: Scalable Universal Image Restoration with Multi-Granularity Degradation Modeling},
author={Jingbo Lin and Zhilu Zhang and Fan Li and Renjing Pei and Wenbo Li and Hang Xu and Hongzhi Zhang and Wangmeng Zuo},
year={2026},
eprint={2611.06042},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2611.06042}
}